The National Center for Health Statistics (NCHS) conducted survey beginning in 1982 on the health, including dental health, and nutrition of Hispanics living in the Southwest, Miami, and New York City. Dental data from the first Southwest component of the survey, referred to as HHANES, will be available for analysis by autumn of 1984. Comprehensive data on the dental health of Hispanics living in the Southwest have already been collected (for description, see Section D). (1) The epidemiologic research team at the University of Michigan's School of Public Health that analyzed the dental data from NHANES I (USPHS/NCHS contract No. 233-79-2092 (2) proposes to analyze the HHANES dental data. Analysis of the data will follow the same approach used during analysis of NHANES I data. After receiving the data tapes, the relevant data will be read, stored, and analyzed using MIDAS computer programs. To control cost of analysis, the more expensive OSIRIS programs, which can allow the findings to be generalized by using sampling weights and by accounting for design effects, will be restricted to selected, pertinent relations identified through MIDAS. Statistical analysis of the data will involve cross-classifying the data by age, gender, income, education, and other potential confounders (using ANOVA, and ANCOVA) and modeling (linear and logistic regression). In the proposed analysis of HHANES' dental data, detailed assessment of the prevalence of dental caries, periodontal disease, tooth loss, and oral hygiene status will be carried out. These dental conditions will also be related to the demographic variables and other risk factors. For example, the relation between consumption of sugary foods and dental caries prevalence will be investigated to see whether the same relations found in NHANES I are also found in Hispanics (2-4). More detailed analysis, such as the relation of periodontal health to smoking and ascorbic acid deficiency, will also be carried out. Comparisons of HHANES and NHANES I will be made to the extent possible within the limitations of data. The excellent computer facilities and statistical support at the University of Michigan will facilitate data analysis. NCHS has expressed interest in publishing the results.